Soil degradation can take many forms, from erosion to salinization to the overall depletion of organic matter. The expression of soil degradation is broad, and so too are the causes. As the world population nears eight billion, and the environmental uncertainty of climate change becomes more manifest, the importance of our soil resources will only increase. The goal of this paper is to synthesize the catalysts of soil degradation and to highlight the interconnected nature of the social and economic causes of soil degradation. An expected three billion people will enter the middle class in the next 20 years; this will lead to an increased demand for meat, dairy products, and consequently grain. As populations rise so do the economic incentives to convert farmland to other purposes. With the intensity and frequency of droughts and flooding increasing, consumer confidence and the ability of crops to reach yield goals are also threatened. In a time of uncertainty, conservation measures are often the first to be sacrificed. In short, we are compromising our soil resources when we need them the most.

Gully erosion is the most severe type of water erosion and is a major land degradation process. Gully erosion susceptibility mapping (GESM)’s efficiency and interpretability remains a challenge, especially in complex terrain areas. In this study, a WoE-MLC model was used to solve the above problem, which combines machine learning classification algorithms and the statistical weight of evidence (WoE) model in the Loess Plateau. The three machine learning (ML) algorithms utilized in this research were random forest (RF), gradient boosted decision trees (GBDT), and extreme gradient boosting (XGBoost). The results showed that: (1) GESM were well predicted by combining both machine learning regression models and WoE-MLC models, with the area under the curve (AUC) values both greater than 0.92, and the latter was more computationally efficient and interpretable; (2) The XGBoost algorithm was more efficient in GESM than the other two algorithms, with the strongest generalization ability and best performance in avoiding overfitting (averaged AUC = 0.947), followed by the RF algorithm (averaged AUC = 0.944), and GBDT algorithm (averaged AUC = 0.938); and (3) slope gradient, land use, and altitude were the main factors for GESM. This study may provide a possible method for gully erosion susceptibility mapping at large scale.

Gully erosion is the destructive and dramatic form of land degradation in Northeast China. The region is the grain production and ecological security base of China where the fertile and productive Mollisols are distributed. Though the region was agriculturally developed relatively recently, it went through high intensity cultivation and fast succession processes within short-time scales. Coupled with irrational farming practice choice and land use, hillslope erosion and gully erosion are seriously threatening agricultural production and environmental stability in the region. The awareness of gully erosion by the local governments started in the 1970s, and conservation measures were thus implemented. In this paper, based on our survey, communications with local farmers and stakeholders as well as investigation for gully erosion for the past three years, we summarize the practical and efficient practices to manage gully erosion developed by researchers and farmers in Northeast China during the past 50 years. These practices include various drop structures, soil check dams, masonry check dams, gabion check dams, wicker check dams, continuous live wicker, a shrub plant enclosure, and an arbor plant enclosure. We specifically expound how a gully erosion practice is set up and identify the site conditions for which they are well-suited. The application of these practices depends on topography, gully size, and local economy. Bioengineering techniques in Northeast China, such as continuous live wicker, a shrub plant enclosure, and an arbor plant enclosure, are highly effective in controlling gully erosion. Problems and challenges are also presented.

Water runoff and sediment transport from agricultural uplands are substantial threats to water quality and sustained crop production. To improve soil and water resources, farmers, conservationists, and policy‐makers must understand how landforms, soil types, farming practices, and rainfall interact with water runoff and soil erosion processes. To that end, the Iowa Daily Erosion Project (IDEP) was designed and implemented in 2003 to inventory these factors across Iowa in the United States. IDEP utilized the Water Erosion Prediction Project (WEPP) soil erosion model along with radar‐derived precipitation data and government‐provided slope, soil, and management information to produce daily estimates of soil erosion and runoff at the township scale (93 km²[36 mi²]). Improved national databases and evolving remote sensing technology now permit the derivation of slope, soil, and field‐level management inputs for WEPP. These remotely sensed parameters, along with more detailed meteorological data, now drive daily WEPP hillslope soil erosion and water runoff estimates at the small watershed scale, approximately 90 km²(35 mi²), across sections of multiple Midwest states. The revisions constitute a substantial improvement as more realistic field conditions are reflected, more detailed weather data are utilized, hill slope sampling density is an order of magnitude greater, and results are aggregated based on surface hydrology enabling further watershed research and analysis. Considering these improvements and the expansion of the project beyond Iowa it was renamed the Daily Erosion Project (DEP). Statistical and comparative evaluations of soil erosion simulations indicate that the sampling density is adequate and the results are defendable. The modeling framework developed is readily adaptable to other regions given suitable inputs. © 2017 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.